Showing papers by "AG Armin Kohlrausch published in 2013"

Binaural Localization and Detection of Speakers in Complex Acoustic Scenes

Abstract: The robust localization of speech sources is required for a wide range of applications, among them hearing aids and teleconferencing systems. This chapter focuses on binaural approaches to estimate the spatial position of multiple competing speakers in adverse acoustic scenarios by only exploiting the signals reaching both ears. A set of experiments is conducted to systematically evaluate the impact of reverberation and interfering noise on speaker-localization performance. In particular, the spatial distribution of the interfering noise has a considerable effect on speaker-localization performance, being most detrimental if the noise field contains strong directional components. In these conditions, interfering noise might be erroneously classified as a speaker position. This observation highlights the necessity to combine the localization stage with a decision about the underlying source type in order to enable a robust localization of speakers in noisy environments.

An Introduction to Binaural Processing

Abstract: The binaural auditory system performs a number of astonishing functions, such as precise localization of sound sources, analysis of auditory scenes, segregation of auditory streams, providing situational awareness in reflective environments, suppression of reverberance, noise and coloration, enhancement of desired talkers over undesired ones, providing spatial impression and the sense of immersion. These functions are of profound interest for technological application and, hence, the subject of increasing engineering efforts. Generic application areas for binaural algorithms are, among others, aural virtual environments, hearing aids, assessment of product-sound quality, room acoustics, speech technology, audio technology, robotic ears, and tools for research into auditory physiology and aural perception. This introductory chapter starts with a discussion of the performance of binaural hearing and then lists relevant areas for technological application. After a short presentation of the physiological background, signal-processing algorithms as applied to binaural modeling are described. These signal-processing algorithms are manifold, but can be roughly divided into localization models and detection models. Both approaches are discussed in some detail. The chapter is meant to serve as an introduction to the main body of the book.

Irrelevant speech effect under stationary and adaptive masking conditions

Abstract: The irrelevant speech effect was investigated in this study where the serial-recall task was performed under six different conditions: Silence, speech-only, noise-only, speech masked by a stationary noise at two different signal-to-noise ratios (SNRs), and speech masked by an adaptive noise. Measured in five test blocks distributed throughout the four test days, the error rate of the serial-recall task under the silence condition sharply decreased in the first few test blocks, halved after completing about seven blocks. When the adaptive masking scheme was used, the error rate of the serial-recall test was reduced compared to the speech-only condition (by 9%) and to the lower-SNR stationary noise (by 4.4%). However, the serial-recall performance was not significantly different between the stationary and the adaptive maskers when the average sound level was carefully matched. Speech Transmission Index (STI) and the correlation coefficient of power spectra were used as the estimators of the temporal and spectral distinctiveness between sound tokens, respectively. The comparison to the test results implied that the frequency-domain estimator may be a better predictor of the relative ISE especially for a non-stationary masker, although it was also suggested that such estimators may have to be combined possibly with an appropriate weighting.

Directional sound masking

Abstract: The invention relates to a system for masking a sound incident on a person. The system comprises a microphone sub-system for capturing the sound. The system further comprises a spectrum-analyzer for determining a power attribute of the sound captured by the multiple microphone sub-system, and a spatial analyzer for determining a directional attribute of the captured sound representative of a direction of incidence on the person. The system further comprises a generator sub-system for generating a masking sound under combined control of the power attribute and the spatial attribute, for masking the incident sound.

Apparent causality affects perceived simultaneity

Abstract: The present research addresses the question of how visual predictive information and implied causality affect audio–visual synchrony perception. Previous research has shown a systematic shift in the likelihood of observers to accept audio-leading stimulus pairs as being apparently simultaneous in variants of audio–visual stimulus pairs that differ in (1) the amount of visual predictive information available and (2) the apparent causal relation between the auditory and visual components. An experiment was designed to separate the predictability and causality explanations, and the results indicated that shifts in subjective simultaneity were explained completely by changes in the implied causal relations in the stimuli and that predictability had no added value. Together with earlier findings, these results further indicate that the observed shifts in subjective simultaneity due to causal relations among auditory and visual events do not reflect a mere change in response strategy, but rather result from early multimodal integration processes in event perception.

Lighting controller, lighting system and lighting control method

Abstract: In a lighting system, sound levels are detected in the vicinity of at least one lighting unit to determine when there is a sound level above a threshold. The illumination provided by the lighting unit is adapted in such a way as to encourage a reduction of noise generation.

Algorithmic Prediction of Inter-song Similarity in Western Popular Music

Abstract: We investigate a method for automatic extraction of inter-song similarity for songs selected from several genres of Western popular music. The specific purpose of this approach is to evaluate the predictive power of different feature extraction sets based on human perception of music similarity and to develop an algorithm able to reproduce and predict human ratings. The algorithm is a linear model that was trained and tested using perceptual data. We use publicly available algorithms to extract acoustic feature values from 78 songs used in a previous perceptual experiment. Feature value differences between songs are used in a multivariate linear regression calculation to find the optimal weighting coefficients for the feature values to best approximate the human similarity perception data. We use two evaluation methods: metrical and ordinal. We use a bootstrapping approach by randomly separating the experimental data into training and testing sets. We compare the performance of this model against...

Source-specific analysis of the noise in an intensive care unit

Abstract: High noise levels in hospitals are often linked to various negative effects on patient outcome and work performance of clinical staff. Despite growing research attention on the adverse acoustic conditions in healthcare environments, few studies offer on-site surveys collected for a relatively long period with a clear description of the measurement protocol, and furthermore, the sources of noise in hospitals are not well documented in the literature. In the current study, the soundscape of an ICU (intensive care unit) room was analysed based on a ~3-day calibrated audio recording, from which acoustic parameters were obtained off-line. In addition, a selected 24-hour recording was annotated, which enabled a source-specific analysis, excluding the patient-generated/-involved contributions. The results showed that the acoustic energy of the noise in this ICU room was attributed to speech and other activities by staff (57%), alarms (30%) and the operational noise of medical devices (13%). In addition, the analysis of the number of loudness peaks showed similar but more uneven proportions: staff (94%), alarm (5%) and device noise (1%). The current study suggests that, to a considerable extent, the noise in ICUs may be attributed to potentially modifiable factors, e.g., staff’s speech and activities.

Off-frequency BMLD : The role of monaural processing

Abstract: Large binaural masking-level differences (BMLDs) can be observed when a tonal signal with an interaural phase difference of π is presented against a diotic masker. The BMLD is large when the signal is spectrally centered in the masker and decreases strongly for off-frequency signals. No such reduction in BMLD would be expected, if monaural detection were governed by energy cues and binaural detection by changes in interaural cross-correlation. The reduction in BMLD thus suggests either that binaural processing is impaired or, alternatively, that additional monaural cues are available in off-frequency conditions. In this study, a stimulus paradigm is used that is expected to impair the processing of additional monaural cues. In the base experiment, a 25-Hz-wide band of diotic noise centered at 700 Hz served as masker. A target tone was presented at 0, 30, 60, and 100 Hz above the masker center frequency, either interaurally in phase (S 0) or out of phase (S π). In the extended experiment, an additional interference tone was always presented spectrally below the masker at the same frequency distance as the target tone was positioned above the masker. The interferer level was 6 dB below the level of the 65 dB masker. By presenting the interferer, a strong modulation is introduced, which should impair the detectability of the target tone based on the beating of masker and target. Results show a small off-frequency BMLD in the base experiment in line with literature. Adding the interference tone produced an increase in both N 0 S 0 and the N 0 S π thresholds, suggesting that monaural modulation cues were indeed used, but also -suggesting that detection performance in the N 0 S π condition was dominated by monaural processing. Additional conditions with modulated interference tones at 500 Hz further supported our hypothesis that monaural modulation cues contributed to reduced -off-frequency BMLDs.

Quantifying temporal ventriloquism in audiovisual synchrony perception.

Abstract: The integration of visual and auditory inputsin the human brain works properly only if the components are perceived in close temporal proximity. In the present study, we quantified cross-modal interactions in the human brain for audiovisual stimuli with temporal asynchronies, using a paradigm from rhythm perception. In this method, partici- pants had to align the temporal position of a target in a rhythmic sequence of four markers. In the first experiment, target and markers consisted of a visual flash or an auditory noise burst, and all four combinations of target and marker modalities were tested. In the same-modality conditions, no temporal biases and a high precision of the adjusted temporal position of the target were observed. In the different-modality conditions, we found a systematic temporal bias of 25-30 ms. In the second part of the first and in a second experiment, we tested conditions in which audiovisual markers with different stimulus onset asynchronies (SOAs) between the two compo- nents and a visual target were used to quantify temporal ventriloquism. The adjusted target positions varied by up to about 50 ms and depended in a systematic way on the SOA and its proximity to the point of subjective synchrony. These data allowed testing different quantitative models. The most satisfying model, based on work by Maij, Brenner, and Smeets (Journal of Neurophysiology 102, 490-495, 2009), linked temporal ventriloquism and the percept of synchrony and was capable of adequately describing the results from the present study, as well as those of some earlier experiments.

Lighting controller and lighting system

Abstract: In a lighting system, sound levels are detected in the vicinity of at least one lighting unit to determine when there is a sound level above a threshold. The illumination provided by the lighting unit is adapted in such a way as to encourage a reduction of noise generation.

Interactive bottle for baby feeding

Abstract: FIELD: medicine.SUBSTANCE: invention relates to medicine. Interactive baby bottle is made with electronic device, which has sensor, configured for measurement of heart rate, and executive mechanism, configured for transmission of measured heart beat to baby. Executive mechanism contains one of multitude of vibration elements, configured for creation of tactile sensation of measured heartbeat, and one or more loudspeakers, located on interactive baby bottle and configured for providing acoustic feedback in real time mode for transmission of measured heartbeat to baby. Vibration element is built into nipple or located in suitable manner on interactive baby bottle for transmission of measured heartbeat to baby via nipple.EFFECT: invention facilitates drinking process, as it makes possible to create sensation of connection between baby and person, feeding baby from bottle, thus approaching breast feeding.13 cl, 6 dwg